Co-registration of clinical magnetic resonance imaging (MRI) and post- mortem brain, together with non-biased stereological cell-counting methods, will be utilized and to determine the pathological substrate of several cortical and subcortical neuroimaging changes in SIVD and AD. Specific Aim 1: Although hippocampal atrophy is an early hallmark of AD, it may also occur in subjects with SIVD. Hippocampal volume (HV), glucose metabolism (CMRglc), and N-acetyl aspartate [NAA] will be correlated with numbers and size of neurons in SIVD and AD. Specific Aim 2: Although SIVD is currently conceptualized as a subcortical disease, new findings during the initial funding period cortical changes (atrophy, decreased CMRglc and [NAA] also occur. Cortical gray matter volume, CMRglc, and [NAA[ will be correlated with number and size of cortical neurons to determine whether a) cortical changes in SIVD represent neuronal shrinkage secondary ti deafferentation, or b) primary neuronal loss due to ischemic injury or other factors. Specific Aim 3: Confluent white matter lesions (WML) are highly prevalent in SIVD, where they are believed to represent areas of myelin and axon loss due to ischemia. The volume and T1-signal intensity of WML will be correlated with numbers of oligodendrocytes, severity of demyelination and axon loss, and severity of microvascular disease. Specific Aim 4: Focal hyperintensities in proton density MRI of the subcortical gray matter (L) are considered to be lacunar infarcts. Because many of these lesions are clinically, the question is posed: Do all hyperintensities represent lacunar infarcts or possibly other types of pathology (e.g., gliosis or incomplete infarction? Specific Aim 3: According to the lacunar hypothesis, lesion location within frontal-subcortical loops is the key determinant of cognitive impairment. Patho-anatomical validation will be sought for a stereotaxic approach used in the Imaging Core to localized lacunes within key individual thalamic nuclei (e.g., anterior and dorsomedial thalamic nuclei. The goal of this project is to enhance our understanding 9of the pathology underlying SIVD and to improve the diagnostic utility of several highly prevalent neuroimaging findings.